rsip4sh Goto Github PK

-glaucoma

Cup and Disk Segmentation of Retinal Images For Detection of Glaucoma Using Deep Learning and Machine Learning(U-NET)

3d-simulation-of-nuclear-fusion-and-fission

The project is an accurate 3D Physics Simulation of Nuclear Fission and Nuclear Fusion. The simulation will show the atomic level of interaction of the particles during Nuclear Fission and Nuclear Fusion. For the simulation on Nuclear Fission, the first fission of the atom to the last formation of the neutrons from the last Nuclear Fission will be displayed. While for Nuclear Fusion, the first fusion of the two particles to the last atom formed will be simulated. The project will be an interactive user-friendly 2D Simulation wherein users will be able to control the number of atoms and the speed of the particles interacting within the simulation. The simulation will be able to be used by students when studying Nuclear Reactions and by teachers when illustrating and simulating Nuclear Reactions.

3dbuildingextraction

this is my master thesis project, which was aimed to extract building models from large 3D urban scenes

3dmv

[ECCV'18] 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation

a-cascaded-r-cnn-with-multiscale-attention-and-imbalanced-samples-for-traffic-sign-detection

a-deeply-supervised-image-fusion-network-for-change-detection-in-remote-sensing-images

Official implement of Paper：A deeply supervised image fusion network for change detection in high resolution bi-temporal remote sening images

abdominal-multi-organ-segmentation

abdominal multi-organ segmentation using pytorch

accuracy-assessment

acsdcf

Adaptive Channel Selection for Robust Visual Tracking with Discriminative Correlation Filters (ACSDCF)

adapted_deep_embeddings

Code associated with "Adapted Deep Embeddings: A Synthesis of Methods for k-Shot Inductive Transfer Learning"

adaptive-filter

Adaptive local noise reduction filter

adaptive-filtering

An implementation of the most common Adaptive Signal Processing Algorithms often used for time-series prediction and noise filtering/cancellation

adaptive_noise_cancel

Implement ANC Filter to recover some very old sound tracks

adaptsegnet

Learning to Adapt Structured Output Space for Semantic Segmentation, CVPR 2018 (spotlight)

adding-noise-and-applying-filters-on-pics

Browse pic and choose to add (Gaussian/salt and pepper) noise or apply (Gaussian/median/average) filter on pic and saving the result of the previous commands in history

advanced-remote-sensing-1

adversarial-perturbations-on-images-and-videos

Code and data from the paper "Targeted Nonlinear Adversarial Perturbations in Images and Videos".

advsemiseg

Adversarial Learning for Semi-supervised Semantic Segmentation, BMVC 2018

aerial-image-superresolution

Dataset used on paper

aerial_wildlife_detection

Tools for detecting wildlife in aerial images using active learning

aerialcrackdetection

Use Keras to detect the pavement cracks in aerial images.

aerialdetection

aes-remote-sensing-images

Soft-threshold Defensive Method for Ad of RSI: Inspired by the Evidences via Throughout Experiments

ag-cnn

The model of "Attention Based Glaucoma Detection: A Large-scale Database with a CNN Model" (CVPR2019)

ai_school_proj

air

Auxiliary Image Regularization for Deep CNNs with Noisy labels at ICLR 2016

albumentations

fast image augmentation library and easy to use wrapper around other libraries

amd-and-glaucoma-detection-using-retinal-funduscopic-image-processin

Image processing is the processing of image by converting an image into digital conformation by applying mathematical operations in the form of signal processing for which the input is an image, a video or a series of image, like video or photograph frame, the outcome of image processing may be either an image or a set of parameters or characteristics related to the image. Glaucoma is the eye disease in which the nerve connecting the eye to the brain is damaged, usually due to high eye pressure. Age-related macular degeneration (AMD) is a decay or breakdown of the eye macula. Glaucoma and age-related macular degeneration (AMD) are these days two of the most incessant reasons for visual impairment and vision misfortune. In addition, high growth of the diseases will be experienced due to diabetes disease incidence increase and the lifestyle lead by populous that is ageing in the present society. Their diagnosis at earliest stage through appropriate good treatment will reduce medical treatment costs generated when they are in early stage else condition may become critical.

an-unsupervised-learning-bidirectional-3d-image-registration

The model contains 1 encoder and 2 decoders for bidirectional (source -> target, target -> source) 3D images registration with skip connections like Unet.

analyzing-the-impact-of-optimization-algorithms-in-training-neural-networks-glaucoma-detection

Glaucoma is one of the leading causes of blindness or vision impairment according to World Health Organization. It requires lot of expertise and practice to detect glaucoma from the retinal color fundus images. To make it scalable and efficient, various machine Learning models have been proposed over the years. Our study is aimed at analyzing the impact of various optimization algorithms and learning rates on training a pre-trained convolutional neural network model for detecting glaucoma from retinal color fundus images. Six different optimizers were studied with seven different learning rates ranging from 0.0005 to 0.1. The Experimental results show that there is a significant impact in Loss and Accuracy of the model with different optimizers and at different learning rates.